how short warm events disrupt snowcover dynamics example ......austre lovén (spitsberg, norvège...

International Snow Science Workshop Grenoble – Chamonix Mont-Blanc - 2013

How short warm events disrupt snowcover dynamics Example of a polar basin – Spitsberg, 79°N

Éric Bernard1, Florian Tolle1, Jean Michel Friedt2, Christelle Marlin3, Madeleine Griselin1 1 CNRS UMR ThéMA, université de Franche-Comté, Besançon, France

2 CNRS FEMTO-ST, université Franche Comté, Besançon, France 3 CNRS UMR IDES, université Paris-Sud, Orsay, France

RESUME : L’Arctique est un indicateur pertinent des changements climatiques. Les dynamiques nivologiques sont directement affectées mais le suivi des processus est rendu délicat par leur rapidité alors qu’ils correspondent à des éléments clef. Ces observations ont pu être conduites sur le glacier Austre Lovén (Spitsberg, Norvège – 79°N) pendant le programme hydro-sensors-FLOWS. Pour mieux comprendre l’impact de ces phénomènes, des observations à échelle spatio-temporelle extrêmement fine ont été menées. Un réseau d’échantillonnage dense (36 points répartis sur 4,5 km2) installé sur le glacier a permis de mesurer différents paramètres : manteau neigeux à plusieurs reprises, bilan de masse, températures au pas de temps horaire. De plus, un réseau de 6 stations photo automatiques permet la cartographie précise du manteau neigeux. Les résultats ont montrés que l’impact de courts événements chauds sur les dynamiques nivo-glaciologiques était capital. La tendance observée ets plus due à un manque de froid l’hiver qu’à un été trop chaud. Un été doux et sec aura une incidence minimale, alors qu’un hiver doux et très arrosé sera extrêmement néfaste aux basses altitudes alors qu’il participera à une importante recharge neigeuse en altitude, au-delà de l’isotherme 0°C. MOTS-CLEF : Changements climatiques, Arctique, dynamiques nivologiques, géomatique. ABSTRACT: Arctic is a reliable indicator of global climate changes. Snow dynamics are directly affected, and following processes becomes challenging: some fast but key events can be missed since they are short but significant. Such observations have been made in the Austre Lovén basin (Spitsbergen, Norway – 79°N) during the Hydro-sensor-FLOWS program. To better understand the impact of climate changes, one approach is to use a fine scale monitoring protocol based on a high density network of sampling points. This has been carried out for 5 years on the Austre Lovén glacier. Overall, 36 measurement points have been distributed at regular intervals over the 4.5 km² glacier surface. For each of these points the snow cover was measured several times and mass balance was yearly recorded. Furthermore, temperature is measured over a 20 sensors network on the glacier, and at last, a network of 6 automated digital cameras is used to map snowcover dynamics. Results show that there is a strong impact of short warm events on snow and ice dynamics. The observed trend is more due to lack of cold in winter than by excess of warm in summer. A meanly warm and dry summer will have few incidence, when a warm and watered winter could be seen as detrimental at low altitude as it will contribute to an important snow refill of the glaciers over the 0°C isotherm. KEYWORDS: Climate change, Arctic, snow dynamics, geomatic. .

1 INTRODUCTION

Climate changes are studied in the Arctic for a long time, but mostly at a large spatio-temporal scale. Several processes observed led us to adopt a scale of observation that is as thin as possible. This is what was done on the small polar glacier Austre Lovén in Spisteberg in the framework of the Hydro-sensor-FLOWS internationnal programm. Results concerning more specifically the impact of seasonal climate

variations on the snowpack will be developed here. The goal is to show that brutal variation, even short, can challenge the results of a complete hydrological year.

2 GEOGRAPHICAL SETTINGS AND FIELDWORK

2.1 Geographical environnement

The Archipelago of Spitsbergen is part of Svalbard, which consists of all islands located between 74° and 80° N and 10° and 35° E. Overall, 60% of the land area is covered by glaciers and it represents about 10% of the total Arctic small glaciers area (Hagen et al., 2003).

Similarly to what is observed in the whole Arctic, it is very reactive to climate change: all

______________________

Éric Bernard, CNRS, Laboratoire ThéMA, Université de Franche-Comté, France; tel: +0033 (0)6 89 73 42 11 email: [email protected]

1226


the small glaciers are clearly retreating since the end of the Little Ice Age (Hagen et al., 2003).

Austre Lovénbreen is a 4.6 km2 alpine type valley glacier located on the west coast of Spitsbergen, on the Brøgger peninsula (79°N, 12°E, Fig. 1), 6 kilometres far from Ny Ålesund settlement. It's a 7 km-long glacier from South to North, and its elevation extends from 50 to 550 m.a.s.l. The glacier area is surrounded by many sharp peaks, whose elevation reaches 876 m.a.s.l (Nobilefjellet). Its hydrological basin represent from about 10 km2. It is a typical small arctic basin which is both exposed to sea and mountain influence.

Figure 1. Localisation and measurement network.

2.2 Climatic condition

The Brøgger peninsula is subject to the influence of the northern extremity of the warm North Atlantic current along the west coast of Spitsbergen (Liestøl, 1993). The present climate at Ny Alesund, along the northern shoreline of the Brøgger peninsula is of polar oceanic climate type. Considering hydrological years (1st October of a given year to 30 September of the next year), the annual average air temperature is -5.77°C (1969-1998) for an average precipitation amount of 391 mm (1969-1998) (source DNMI at http://eklima.met.no). Over the period 1969 to 2010, these parameters display values increasing with time: +0.51°C per decade in temperature and +14 mm per decade in precipitation. However, the time-series may be separated into two distinct periods, which indicates a significant climate change at least from the late 90s.

During the first 29 years (1969-1998), the mean value was -5.77°C. There was no significant temporal gradient (-0.02°C per

decade) in comparison with the last 12 years (1998-2010).

From 1998 until 2010, the air temperature significantly increased with time (annual gradient of +1.26°C per decade). This recent gradient is around 2.5 times the average one calculated for the whole period (1969-2010). The mean air temperature during this 12-years period is always above the 1969-1998 average value that is always above -5.77°C (Figure 2). The mean air temperature value is -4.20°C for 1998 to 2010 period.

3 METHODS AND DATA

A specific measurements network was implemented on the Austre Lovén glacier, so as to follow the hydro-nivo-glaciological process (Fig. 2). This network has been designed to provide the most accurate spatio-temporal survey.

To complement satellite images, 6 automated digital camera-based have been installed. This setup, thanks to which oblique view pictures are gathered with high time and spatial resolution (the pixel size is at worst 0.46 m X 0.46 m considering the lens aperture), covers 96% of the glacier surface. We have selected the acquisition of 3 images every day at a fixed hour set to 7h, 11h and 15h UTC (corresponding to 8/12/16h (mean solar time at this longitude). Hence, weather permitting, a daily coverage of the glacier condition is guaranteed by this instrument network, providing a perfect complement to spaceborne satellite imagery whose availability is strongly dependent on weather conditions and satellite programming over the region of interest.

1227


Figure 2. Monitoring on the Austre Lovén glacier.

A network of 20 air temperature logger is distributed over the whole surface of the glacier (Fig. 1). This network gives the hourly temperature during all the year, on each of these points. This allows to calculate the thermal state of the glacier (hourly, daily or monthly depending on process studied) both giving a value for the whole surface and spatializing the information by interpolating data (Bernard et al., 2011). This data processing gives the possibility of observing each warm event, whatever their duration is. In addition, we can determine which part of the glacier is concerned by those specific events.

Concerning the precipitations, data used are those of Ny Alsund meteorological station, 6 km far from the glacier. Sensors on the glacier do not yield good results because of field conditions.

4 RESULTS

4.1 First observations

Results show that there is a strong impact of short warm events on snow and ice dynamics. Indeed, the last 5 years exhibit some clues to understand some key processes.

This short period belong to a hot decade which affects Svalbard: the observed trend is more due to lack of cold in winter than by excess of warm in summer. Observations at a

local scale show the main impact of liquid precipitations on snowcover dynamics and then on glacier’s behaviour each season. The result is complex: a meanly warm and dry summer will have few incidence on the glacier mass balance, when a warm and very dry winter could be seen as detrimental at low altitude as it will contribute to an important snow refill of the glaciers over the 0°C isotherm.

4.2 Glacier thermal state

We note that, with the exception of warm events, there is no positive thermal state before May 1st. Regarding the thermal state of the glacier, a day can display both positive and negative temperatures, both temporally as spatially during all the year. These alternations continue late in the season, and full positive thermal state (on a daily scale of the glacier), does really install only during the first half of June.

Figure 3. Thermal state of the Austre Lovén glacier.

In 2008-2009, the average of the thermal states reach - 6.96 °C, against - 5.06 °C in 2009-2010. However, we note that the melting

1228


season is in average warmer in 2008-2009 (0.8 °C) than in 2009-2010 (0.6 °C).

If it is difficult to identify clear trends, however, we note that the thermal summer, with temperatures significantly and sustainably positive, usually begins in mid-June to end during the first week of September.

4.3 Importance of winters

Winter is characterized by the occurrence of warm events: accompanied by precipitation, they have a major impact on the snowcover and therefore on glacier’s accumulation.

Winter 2008-2009 shows a significant temperature oscillation with many very cold peaks, reaching almost the - 30 °C in January 2009. Between January and the end of May, when the thermal state rises very sharply (up from -20 °C to +2 °C), there are five peaks with minimum located between - 20 °C - 25 °C. This winter is thus punctuated with six warm events, resulting from a sharp rise in temperatures. For each of these events, the period above 0 °C is extremely short (1.5 days on average), and temperatures drop as quickly as they are rose. It is an outstanding phenomenon, since the amplitudes of the thermal states are always above 15 °C.

Winter 2009-2010 has similarities with the previous season. The most important cold peak is also involved January. However, this winter is less cold, but mostly warm events are much more counted. Between November and May, there are indeed nine warm events that follow each time an extremely sharp rise in temperatures.

The thermal behaviour of the winter is not linear, but marked by extreme fluctuations in temperature. This could be nothing if it appears during dry conditions, unlike precipitations occur.

5 DISCUSSION

5.1 Brutality of temperature fluctuations

To demonstrate this particularity, we can take the example of November 26th 2009 (Fig. 4). The details of the hourly evolution of the thermal state reveals fluctuations in the 0 °C isotherm.

Figure 4. Daily thermal state and its Daily temperature sudden variation.

It is thus noted that the glacier was left

entirely in positive values for nearly two hours, which is sufficient to initiate melting dynamics. Such event, which creates circulation of liquid water within the snowcover, is causing crusts found in the stratigraphic profiles. In addition, melt accompanied by rain greatly reduces mass and thickness of the snowcover.

5.2 Winter short warm events

Let’s take the example of one warm event that occurred on January 15, 2010. This one was particularly representative.

Although the daily thermal states of January 15 and 16 are negative (about -0.6 °C), there is a relatively large area of the glacier (22 and 24%) in positive values. On this part of the glacier, heavy precipitations are probably rainfall, with consequences on snowcover, which could completely disappear. Following the refreezing observed on the following days, a large amount of the precipitation could be found in the form of melt-freeze crust in the snowcover.

1229


Figure 5. Daily thermal state and its Daily temperature sudden variation.

5.3 Visual results

According to the camera network, the figure 6 shows one of the many impacts of warm events accompanied by rain. This was observed at the late October, when winter seemed to be installed. It took just a few hours of huge rainfall occurred simultaneously with an important rise of temperature to destroy completely the snowcover. This warm event lasted less than a day: thermal state was significantly negative before and after this event.

Figure 6. Impact of a warm event on the snowcover when accompanied by rainfall. On the left, the situation on October 18th and on the left one day later.

6 CONCLUSION

Despite of a general climatic trend along an hydrological year, short warm event observed and described above could show reversing effects on snowcover over a period of only a few days or sometimes a few hours.

7 REFERENCES

Bernard E., Friedt JM., Tolle F., Griselin M., Laffly D. et Marlin Ch., 2013 : « Seasonal snow and ice dynamics monitoring using ground based high resolution photography (Austre Lovénbreen, Svalbard, 79°N) », ISPRS Journal of Photogrammetry and Remote Sensing, N°75, pp 92-100.

Bernard E., Tolle F., Griselin M. et Marlin Ch., 2011 : Quantification des hauteurs de neige et des températures de l’air à la surface d’un glacier : du terrain à l’interpolation, confrontation de méthodes, Actes des IXe Rencontres de ThéoQuant 2009, 11 p.

Bednorz E., 2011 : Occurrence of winter air temperature extremes in Central Spitsbergen, Theoretical and Applied Climatology, consultation en ligne.

Escher-Vetter H. & Siebers M., 2007 : Sensitivity of glacier runoff to summer snowfall events, Annals of Glaciology, Vol. 46.

Farinotti, D., Magnusson, J., Huss, M., Bauder, A., 2010. Snow accumulation distribution inferred from time-lapse photography and simple modelling. Hydrological Processes 24 (15), 2087–2097.

Hagen, J. O., Kohler, J., Melvold, K., Winther, J.-G., 2003. Glaciers in Svalbard: mass balance, runoff and freshwater flux. Polar Research 22,145{159.

1230


Hinkler, J., Pedersen, S., Rasch, M., Hansen, B.U., 2002. Automatic snow cover monitoring at high temporal and spatial resolution, using images taken by a standard digital camera. International Journal of Remote Sensing 23 (21), 4669–4682.

Humlum, O., 2004. Mapping snow cover duration, avalanches and other geomorphic processes by automatic digital cameras, Longyeardalen Svalbard – a project funded by the University Courses on Svalbard (UNIS) 2000–2005. UNIS, Department of Geology, Svalbard. <http://www.unis.no/research/ geology/Geo_research/Ole/mapping_snow_cover_duration.htm>.

Kholer J., Brandt O., Johansson M., & Callaghan T., 2006 : “A long-term Arctic snow depth record from Abisko, northern Sweden, 1913–2004”, Polar Research 25 (2), pp 91-113.

Mernild S.H. & Liston G.E., 2010 : The influence of air temperature inversions on snow melt and glacier surface mass-balance simulations, Ammassalik Island, Southeast Greenland. Journal of App.lied Meteorology and Climatology, Vol. 49, No. 1, pp. 47–67.

Oerlemans E., 2005 : Extracting a climate signal from 169 glacier records, Science, No. 308, pp. 675-677.

Przybylak R., 2007 : Recent air-temperature changes in the Arctic, Annals of Glaciology, Vol. 46, No. 1, pp. 316-324.

Tveito O. & Schöner W., 2002 : Applications of spatial interpolation of climatological and meteorological elements by the use of geographical information systems (GIS). Cost 719, Rep. 28/02. Norwegian. Meteorologisk. Institutt, Blindern, Oslo, Norge, 44 p.

1231

how short warm events disrupt snowcover dynamics example ......austre lovén (spitsberg, norvège...

Documents